All packsSaaS Billing + Tax CoreSolve
stripe customer resubscribe new subscription id cache bug
A returning customer gets a new subscription id
Keying a local billing cache by Stripe subscription id breaks the moment a canceled customer resubscribes and gets a brand-new sub_ id (D2).
This is one of the things SaaS Billing + Tax Core already handles. Charge per-seat subscriptions: upgrades, downgrades, failed cards, invoices and sales tax.
Is this you?
A customer who cancels and later resubscribes gets an entirely new Stripe subscription id, not a reactivated old one. A local cache table keyed by that subscription id either collides with the old canceled row or silently forks into two rows for one account -- either way, the app's idea of "is this account paying" goes wrong for a real, common lifecycle event, not an edge case.
Why this one is easy to get wrong
Keying the cache by the Stripe object's own id looks like the obvious, safe choice -- it's unique, it's what Stripe gives you first. The bug only appears on the second subscription for the same account, which most manual testing never exercises because it means canceling and waiting, not just checking out once.
What you get instead
D2 keys the Subscription cache row by accountId, not by Stripe subscription id, specifically so a churn-and-return customer updates the same row instead of orphaning it. D5's supersession rules then govern exactly when a different incoming subscription id is allowed to replace that row's contents versus being refused as a logged, operator-visible orphan.
Source: ARCHITECTURE.md D2 — checkable in the pack you receive
How you actually use this
You don’t install a library or wire up an SDK. Your own coding agent builds the code in your project, and you keep it — no runtime dependency on us.
Step 1
Download and unzip
You get a folder: the docs that tell an agent what to build, a starting skeleton, and the test suite that decides when it's done.
Step 2
Open it in Claude Code or Cursor
Point your coding agent at the folder. Nothing to install, no account with us, no API key.
Step 3
Paste one prompt
The pack contains the exact prompt. Paste it as your first message and leave it alone — it works through the build itself, choosing a cheaper or stronger model per task.
Step 4
Run ./verify.sh
One command. It prints a pass or fail for every check. Green means the build is done — the same script we ran to produce the receipt on this page.
Typical build: about 18 minutes of your agent working, mostly unattended. Then you integrate the working module into your app the way you would any code you’d written yourself.
Why you can believe this
3 of 3 runs passedWe ran this pack from an empty folder 3 times and published exactly what happened — every check, the model, the token cost, the wall time. Not a testimonial, and not our opinion: the same verify.sh you run yourself. Read the full receipt →